DRAGON Documentation

DRAGON, for DiRected Acyclic Graphs OptimizatioN, is an open source Python package for the optimization of Deep Neural Networks Hyperparameters and Architecture [1]. DRAGON is based on the package Zellij.

The distributed version requires a MPI library, such as MPICH or Open MPI. It is based on mpi4py.

Install DRAGON

Basic version

After cloning the git repository, install DRAGON, using:

pip install -e dragon

Distributed version

If you plan on using the distributed version, you have to install the mpi4py package:

pip install mpi4py

Dependencies

• Python >=3.9

• numpy>=1.21.4

• DEAP>=1.3.1

• botorch>=0.6.3.1

• gluonts>=0.11.3

• gpytorch>=1.6.0

• graphviz>=0.8.4

• pandas>=1.3.4

• enlighten>=1.10.2

• [mpi]: mpi4py>=3.1.2

Contributors

• Julie Keisler: julie.keisler.rfo@gmail.com

References

[1]

Keisler, J., Talbi, E. G., Claudel, S., & Cabriel, G. (2023). An algorithmic framework for the optimization of deep neural networks architectures and hyperparameters. arXiv preprint arXiv:2303.12797.